EN中文

Journal of Resources and Ecology >

2024 , Vol. 15 >Issue 4: 898 - 908

DOI: https://doi.org/10.5814/j.issn.1674-764x.2024.04.011

Impact of Human Activities on Ecosystem

How Does the Blue C Stock Vary in Restored and Degraded Wetlands Across Land Cover Mosaics? Evidences from Medinipur Coastal Plain, India

  • Mansa DEY , 1 ,
  • Mrinmoyee NASKAR 1, 2 ,
  • Sohini NEOGY 3 ,
  • Debajit DATTA , 1, *
Expand
  • 1. Landscape Ecology Laboratory, Department of Geography, Jadavpur University, Kolkata 700032, India
  • 2. Department of Geography, Baruipur College, Baruipur 743610, India
  • 3. Department of Geography, Banwarilal Bhalotia College, Asansol 713303, India
* Debajit DATTA, E-mail:

Mansa DEY, E-mail:

Received date: 2023-10-03

  Accepted date: 2024-01-20

  Online published: 2024-07-25

Supported by

The Early Career Research Award(ECR/2017/003380)

The Project of Science and Engineering Research Board, Department of Science and Technology, Government of India(DST-SERB)

Fold

Abstract

Coastal wetlands are of paramount importance as major reservoirs of blue carbon (C), playing a crucial role in providing nature-based solutions to mitigate climatic changes. This research aimed to analyse the dynamics of total blue C (TBC) and its components; viz. soil organic C (SOC), below ground C, and above ground C; as well as how they are influenced by land use/ land cover (LULC) categories and wetland situations. Subsequently, study sites were identified as one restored wetland and another degraded wetland in the Medinipur Coastal Plain, India. The LULC categories were analyzed using Pleiades 1A and 1B satellite imagery, corresponding to the restored and degraded wetland, respectively. The quantification of SOC was based on point-specific sample data collected from both wetlands (nr=250; nd=84). Above ground biomass (AGB) was appraised employing allometric relationships involving field-measured dendrometric variables. Below ground biomass values were calculated using indirect allometric equations that take into account the AGB values. Integrating all the components, TBC stock of the restored and degraded wetlands were estimated at 246710.91 Mg and 7865.49 Mg, respectively. In the restored wetland, dense mangrove and open mangrove exhibited higher concentrations of blue C components, while other LULC categories demonstrated moderate to low densities. In the degraded wetland, the open mangrove category recorded high densities of C pools, whereas herbaceous vegetation, bare earth and sand, and waterbody exhibited lower concentrations. The results portrayed significant disparities (P<0.05) in blue C pools among different LULC categories in both wetlands. Furthermore, it was evident that wetland type and LULC category had notable (P<0.001) impacts on TBC dynamics, both individually and in combination. Overall, this research may aid in effective management of coastal wetlands as blue C sinks, emphasizing their significance as essential elements of climate change mitigation strategies.

Key words： blue C pools; coastal wetland; dendrometric variable; landscape transformation; mangrove plantation; wetland restoration

Cite this article

Mansa DEY , Mrinmoyee NASKAR , Sohini NEOGY , Debajit DATTA . How Does the Blue C Stock Vary in Restored and Degraded Wetlands Across Land Cover Mosaics? Evidences from Medinipur Coastal Plain, India[J]. Journal of Resources and Ecology, 2024 , 15(4) : 898 -908 . DOI: 10.5814/j.issn.1674-764x.2024.04.011

1 Introduction

The ‘blue Carbon (C)’ can be captured and accumulated by coastal ecosystems and may stay there for centuries (Hamzeh and Lahijani, 2022; Datta et al., 2023). Blue Crefers to the entire amount of organic C (Corg) that is stored in tidal marshes, mangrove patches, mudflats, and seagrass beds as soil organic C (SOC), below ground C (BGC), and above ground C (AGC) (Xi et al., 2021; Datta et al., 2023). All these Corg reserves together yield the total blue C (TBC) (UNEP and CIFOR, 2014). In addition to offering evident ecological services, these are also critical for human well- being as well as biodiversity conservation (Sutton-Grier and Sandifer, 2019). However, coastal blue C pools are one of the most vulnerable ecosystems on earth and are experiencing significant depletion year after year (McLeod et al., 2011). Unfortunately, the depletion of these precious natural areas contributes to increased CO2 emissions, which further aggravates climate change (Pendleton et al., 2012).
Research on coastal blue C has received substantial interest in recent decades. Huge amounts of work have been put into component-specific studies globally (Lu, 2005; Haywood et al., 2020). Dendrometric variables and remote sensing (RS) data were employed for biomass appraisal (both above and below ground), either separately or in combination (Deb et al., 2017; Kenzo et al., 2023). Furthermore, to increase the precision of estimating the amount of C sequestered by vegetation, biomass has been projected by modeling techniques and/or analysis of the effectiveness of machine learning methods (Zhou et al., 2023). Generally, multi-sensory satellite datasets have been utilized to estimate biomass (Luo et al., 2017; Bazzo et al., 2023). Few investigations on SOC estimation have concentrated on point-specific observations; nonetheless, works of Vågen et al. (2013), Datta et al. (2022), and Yang et al. (2023) sought to spatially explicitly estimate the total SOC stock. Although the role of LULCs in altering coastal wetland environs has been studied globally, only a handful of researches have attempted to inspect the influence of LULC patterns on TBC dynamics (Ma et al., 2019; Wu et al., 2023).
India, as a geographically diverse country, bolsters a plethora of unique wetland habitats, which, in turn, delivers an extensive array of ecosystem services and products (Bassi et al., 2014; Mohanta et al., 2020). Multitude researches have been carried out in India concerning TBC stocks (Das et al., 2021; Babu et al., 2023; Datta et al., 2023). Nevertheless, considering the heterogeneous Indian landscape and the proliferating anthropogenic stressors, very limited researches have emphasized the influence of LULC patterns on blue C storage except those of Bhomia et al. (2016) and Rani et al. (2021). In an exceedingly populated country like India, where anthropogenic activities regularly alter, transform, and degrade the wetland and coastal landscapes, the issue of recognizing LULC patterns as one of the primary drivers of blue C dynamics becomes more pressing (Perera et al., 2022). Numerous wetland restoration initiatives have been put in place to improve C storage and vegetation structure to reduce wetland deterioration (Dung et al., 2016). Innovative pecuniary incentives are being developed, particularly in emerging economies, to promote local and national level forest conservation (Senger et al., 2021). Sustainable strategies like Reducing Emissions from Deforestation and Degradation (REDD+) are currently introduced to calculate net carbon reductions. It is therefore crucial to estimate the TBC stocks within restored and degraded mangroves.
Considering the current scenario, the present study seeks to evaluate 1) how LULC category and wetland type influence the TBC dynamics; and 2) a comparative analysis of the distribution patterns of TBC and its components (AGB, BGB, and SOC) across various LULC classes within both restored and degraded wetlands. In this regard, we deliberately chose one restored and one degraded wetlands located along the Medinipur Coastal Plain (MCP), India as our designated case study sites.

2 Methods

2.1 Study sites

This research emphasized two characteristically different wetland sites located in the MCP, viz. restored wetland and degraded wetland. The Bichitrapur Mangrove Sanctuary (restored wetland), is situated at the confluence of the river Subarnarekha in the state of Odisha, India. The studied restored wetland encompasses nearly 1350.52 ha (Fig. 1a). Historically, this area featured a substantial natural mangrove population up to the late 1970s (Datta et al., 2022). Nevertheless, owing to the combined impacts of anthropogenic interferences and recurrent natural hazards, this ecosystem faced severe degradation. In response to this challenge, the forest department of Odisha demarcated this expanse as a Proposed Reserve Forest. After receiving support from the Odisha Forestry Sector Development Project (OFSDP), they initiated community-based awareness and plantation programs to protect, restore, and rejuvenate the mangrove forest (OFSDP, 2010; Roy and Datta, 2018). Currently, this restored wetland is thriving with various mangrove and mangrove-associated species, including Avicennia alba, Avicennia marina, Bruguiera gymnorhiza, Acanthus ilicifolius, Excoecaria agallocha, and more (Datta et al., 2022).
Fig. 1 Locations of the (a) restored and (b) degraded wetland with corresponding in-situ sampling plots
Conversely, the degraded wetland is located at Dadanpatra of MCP, in the state of West Bengal, India. This degraded wetland encompasses an area of approximately 84.55 ha (Fig. 1b). Notably, this specific wetland is categorized as a salt marsh (Roy et al., 2020). However, rapid urbanization, the growth of tourism leading to extensive infrastructure development, the expansion of commercial brackish water aquaculture, unsustainable exploitation of natural resources, improper land management practices, inadequate conservation efforts, lack of enforcement of existing regulations, and a deficiency in community awareness has resulted in degradation of ecosystem health in this wetland site (Datta et al., 2022). Specifically, the conversion of natural land covers for commercial brackish water aquaculture and salt farming stands out as a major contributor to the deterioration of this wetland.

2.2 LULC mapping

Cloud-free Pleiades 1A satellite image, acquired on 25th November, 2018 was used for the study of the restored wetland (Supplementary Table 1). Alternatively, the degraded wetland was studied by using cloud-free Pleiades 1B satellite image (spatial resolution: 2 m), which was acquired on 28th January, 2020. Provided by Airbus Defence and Space, both the images are orthorectified and having multispectral bands (blue, green, red, near-infrared). These datasets exhibited lesser likelihood of making misclassification errors during the spectral analysis of various LULC types because they are typically from the same season (Pena and Brenning, 2015). Researchers worldwide have utilized a range of approaches to classify remote sensing datasets, resulting in differing assessments regarding the most efficient classification method (Datta et al., 2021). Nevertheless, several researches have highlighted the efficacy of the Support Vector Machine (SVM) classifier regarding supervised classification of RS database (Dash et al., 2023). Accordingly, SVM technique was applied for identifying nine LULC categories for the restored wetland, namely, dense mangrove, open mangrove, mixed vegetation, Casuarina plantation, herbaceous vegetation, agriculture, aquaculture, waterbody, and bare earth and sand; and four LULC classes for the degraded wetland, which includes bare earth and sand, open mangrove, herbaceous vegetation, and waterbody. Using ENVI 5.3 software, detailed LULC maps were generated by integrating field-based observations, Pleiades data, and Google Earth-derived data into the SVM classifier (Datta et al., 2022).
Supplementary Table 1 Detailed specifications of the dataset used in this study
Dataset Data type Specifications
Pleiades 1A and 1B Multispectral dataset Blue (0.43-0.55 µm)
Green (0.50-0.62 µm)
Red (0.59-0.71 µm)
Near Infrared (0.74-0.94 µm)
PALSAR-2 Synthetic aperture radar (SAR) dataset L-band (HH and HV)
Wood density Secondary dataset Global compilation of region and species specific wood density

2.3 Accuracy assessment

Following the SVM algorithm-based supervised classification of existing LULCs, classification accuracy assessments were conducted to quantitatively evaluate the accuracy of assigning pixels to specific LULC classes (Jia et al., 2014). The locational database of all recognized LULC classes had been compiled from 192 and 67 in-situ sampling points corresponding to the restored and degraded wetlands, respectively. Subsequently, these locations were utilized as references to evaluate classification precision by computing overall accuracy and kappa coefficient.

2.4 Sampling procedure for TBC estimation

In this study, a stratified random sampling technique has been adopted to ensure the comprehensive representation of all LULC categories present within the wetland sites. It was also taken into account that sampling proportions closely aligned with the respective proportions of LULC types within the wetlands.
Using the SOC dataset generated by Datta et al. (2022), point-specific SOC stocks of the restored wetland were calculated. The same methodology was followed for calculating the SOC storage of degraded site. A Russian Sediment/ Peat Borer was used to collect 89 and 25 soil cores, in the two wetlands respectively. Cores were collected down to 1 m depth in four successive depth intervals (depth 1: 0-20 cm; depth 2: 20-40 cm; depth 3: 40-70 cm; depth 4: 70-100 cm). Subsequently, the soil samples (nr=250; nd=84) were tested for SOC estimation using a procedure after Kauffman and Donato (2012) and Fourqurean et al. (2014). Continuous sand and gravel beds hindered the acquisition of soil cores for all the successive depths in several LULC categories. The dataset adequately covered all of the existing LULC kinds proportionately.
Both destructive and non-destructive sampling processes were employed for the estimation of AGB. A quadrat survey method was employed between November, 2018 and March, 2019 for the field-based ecological measurements. At every 10 m×10 m sampling plot, the tree-AGB was determined using several dendrometric variables and conventional allometric formulae. The trees that fell within a sample plot had their girth at breast height assessed by a measuring tape. The diameter (D) of tree trunks were derived from these girth measurements. Additionally, wood density (ρ) was acquired from the World Agroforestry database (World Agroforestry Centre, 2011). Height of a tree (H) was estimated using a laser distance meter (Leica® DISTROTM D-510, Switzerland). The AGB of a specific tree was computed following Chave et al. (2005), while taking into account the eco-climatic condition corresponding to the particular species (Peichl and Arain, 2007).
$AG{{B}_{\text{dry}}}=0.112\times {{(\rho \times {{D}^{2}}\times H)}^{0.916}}$
$AG{{B}_{\text{moist}}}=0.0509\times (\rho \times {{D}^{2}}\times H)$
where, AGBdry/moist denotes assessed AGB of dry/moist tree (kg), D denotes trunk diameter (cm), H denotes total tree height (m), and $\rho $ denotes wood specific gravity (g cm–3).
Within the original plots, the AGB of shrubs was estimated in 1 m2 sized sub-plots. The following equation had been used to calculate stand-specific estimations of shrub-AGB (Ali et al., 2015).
$\ln AGB=-3.50+1.65\times \ln D+0.842\times \ln H$
where, D is the longest stem diameter (cm), and H is shrub height (m).
Destructive sampling was conducted in the smallest plots of 2500 cm2 size within the designated subplots mentioned above to appraise the AGB of ground litter, herbaceous plants, pneumatophores etc. (Cummings et al., 2002). Laboratory tests were conducted on the materials involving oven drying and gravimetric techniques. The line transect (diagonal lines inside the main plots) technique was applied for non-destructive sampling of the biomass of downed wood and to calculate its AGB (Kauffman and Cole, 2010). Minor portions of wood were viewed as litter. To determine the AGB of dead trees in the main plots, the method of Manuri et al. (2014) was adjusted as per the present scenario and used accordingly. By subtracting portions (in general, 2.5% of the AGB of a tree) lost due to missing components, adjustments were required to fit the tree's current state (Kauffman and Donato, 2012). Additionally, destructive sampling had been used to estimate the AGB of aquatic vegetation in 1 m2 buoyant PVC quadrats that served as plots in the watered sections of the study sites (Madsen, 1993). Accumulated samples were analyzed using oven-drying and weighing procedures in an ex-situ setup. Consequently, plot level AGB was calculated and translated into globally accepted Mg ha-1 unit by summing all the AGB constituents detected in a particular plot (Kauffman and Donato, 2012). Indirect allometric equations incorporating the AGB database were applied to calculate BGB (Hutchison et al., 2014). These were (Luo et al., 2012):
In case of mangrove species:
$BGB=0.073\times AG{{B}^{1.32}}$
In case of non-mangrove species:
$BGB=1{{0}^{\left( \left( 0.949\times lo{{g}_{10}}\left( AGB \right) \right)-0.531 \right)}}$
Thereafter, using specific conversion factors as mentioned by Kauffman and Donato (2012), the AGB and BGB datasets were transformed to above ground and below ground C (AGC and BGC). Subsequently, by summing up these already acquired AGC, BGC, and SOC values, the TBC dataset had been generated using the following equation:
$TBC=AGB+BGB+SOC$
The spatial explicit mapping of the TBC stock estimation for both wetlands was carried out following the method after Datta et al. (2023).

2.5 Analysis of relationships between LULC, wetland type, and TBC components

The impact of LULC categories, and wetland type on the distribution pattern of TBC density, including its components, viz. AGB, BGB, and SOC were analyzed through statistical analyses. Mean (μ) and standard deviation (σ) were calculated for analyzing LULC-specific distributions of AGB, BGB, SOC, and TBC densities. The intra-wetland and inter-wetland comparisons were conducted to gain insights into how various LULC classes influence the C stock dynamics of a particular wetland. Additionally, we sought to understand how a specific LULC category exerted control over the dynamics of C stock in different types of wetlands. Subsequently, one-way analysis of variance (ANOVA) had been employed for investigating variations in average densities of TBC as well as its components across LULC categories (Datta et al., 2022). When ANOVA revealed statistically significant variations among LULC categories, Tukey’s Honestly Significant Difference (HSD) analysis (two-tailed) was employed for identifying specific variations among LULC classes (P<0.05) (Nandi et al., 2020). Furthermore, two-way ANOVA was carried out for assessing the individual and combined influences of wetland type as well as LULC classes on the mean AGB, BGB, SOC, and TBC densities (Datta and Deb, 2017).

3 Results

3.1 LULC pattern of selected wetlands

The restored and degraded wetlands had been categorized into nine and four LULC classes, respectively, on the basis of ground scenario (Fig. 2). The entire restored wetland was dominated by herbaceous vegetation (23.16%), dense mangrove (22.71%), and open mangrove (19.51%) classes, re-spectively. Contrarily, agriculture (4.03%), aquaculture (3.55%), mixed vegetation (5.03%), and bare earth and sand (3.36%) were observed as the LULC classes with the lowest amount of area coverage. Meanwhile, a moderate amount of areal coverage was observed for the Casuarina plantation (9.71%) and water body (8.94%) categories. In the case of degraded wetlands, herbaceous vegetation was the predominant class, covering 42.48% of the total area, followed by bare earth and sand (33.30%) and waterbody (23.34%) categories. These LULC classes were found to be sparsely scattered throughout the entire wetland. Conversely, a concentration of a few patches of open mangrove (0.88%) was observed on the southern part of the wetland, covering the lowest amount of area coverage. The generated LULC maps indicated an overall accuracy of 92.19% for the restored wetland and 92.54% for the degraded wetland.
Fig. 2 LULC categories of the (a) restored and (b) degraded wetlands

3.2 TBC dynamics of the wetlands

The TBC dynamics in the restored wetland depicted extensive fluctuations with values varying between 0.34 Mg ha-1 and 881.50 Mg ha-1. Overall, highest TBC values were concentrated in the central area. It gradually decreased outward in a radiating pattern (Fig. 3a). Notably, the dense mangrove exhibited exceptionally high TBC density (μ: 412.96 Mg ha-1, σ: ±153.39 Mg ha-1), while the open mangrove showed moderately high values (μ: 182.98 Mg ha-1, σ: ±108.29 Mg ha-1). Conversely, the Casuarina plantations displayed the lowest TBC density (μ: 67.00 Mg ha-1, σ: ± 60.51 Mg ha-1). The appraised TBC storage of the restored wetland was 246710.91 Mg (Table 1). The dense mangrove (126667.22 Mg) stood out as the largest TBC sink of the wetland, followed by open mangrove (48200.59 Mg), herbaceous vegetation (31003.59 Mg), and waterbody (13592.75 Mg). Conversely, the LULC category of bare earth and sand (3525.77 Mg) exhibited lowest TBC stock, followed by aquaculture (3597.63 Mg) and agriculture (4611.85 Mg). It is noteworthy that the mean TBC densities among various LULC classes demonstrated significant deviations (P<0.05), with exceptions for mixed vegetation in relation to herbaceous vegetation as well as aquaculture in relation to bare earth and sand.
Fig. 3 Spatial distribution of estimated TBC stocks of the (a) restored and (b) degraded wetlands

Note: The TBC dynamics of the degraded wetland displayed a comparatively narrow range, that is from 2.50 Mg ha-1 to 279.23 Mg ha-1 (Fig. 3b). Distinct high values were found at the central portion of the wetland which seems to gradually disperse outward following a decreasing trend. Moderate TBC density was recorded in bare earth and sand (μ: 108.48 Mg ha-1, σ: ±34.98 Mg ha-1) and open mangrove (μ: 102.66 Mg ha-1, σ: ±42.87 Mg ha-1), while lower TBC density was observed in herbaceous vegetation (μ: 83.65 Mg ha-1, σ: ±41.26 Mg ha-1) and waterbody (μ: 87.64 Mg ha-1, σ: ±39.27 Mg ha-1).

Table 1 Estimated TBC stocks, mean (μ), and standard deviation (σ) of TBC densities under different LULCs
Wetland LULC category Area (ha) TBC density (Mg ha-1) TBC stock (Mg) Total TBC stock (Mg)
μ±σ
Restored Dense mangrove 306.73 412.96±153.39A 126667.22 246710.91
Open mangrove 263.42 182.98±108.29B 48200.59
Mixed vegetation 67.89 99.00±83.49C 6721.11
Casuarina plantation 131.20 67.00±60.51D 8790.40
Herbaceous vegetation 312.82 99.11±78.87C 31003.59
Agriculture 54.43 84.73±67.99E 4611.85
Aquaculture 47.93 75.06±73.56F 3597.63
Bare earth and sand 45.33 77.78±68.70F 3525.77
Waterbody 120.76 112.56±114.92G 13592.75
Degraded Open mangrove 0.74 102.66±42.87A 75.97 7865.49
Herbaceous vegetation 35.92 83.65±41.26B 3004.71
Bare earth and sand 28.16 108.48±34.98C 3054.80
Waterbody 19.74 87.64±39.27D 1730.01

Note: Diverse superscript letters show significant deviations in TBC densities across LULC categories as per one-way ANOVA (F) first and thereafter through Tukey’s HSD Post Hoc measure (P<0.05). Degrees of freedom (df) are (8, 3376282) for restored and (3, 211381) for degraded wetlands, respectively.

The estimated TBC stock of the degraded wetland under investigation was 7865.49 Mg. Among all LULCs, the bare earth and sand (3054.80 Mg) recorded the largest TBC stock, followed by herbaceous vegetation (3004.71 Mg), waterbody (1730.01 Mg), and open mangrove (75.97 Mg). The mean TBC density under various LULC classes of the degraded wetland significantly differed (P<0.05) from each other. Furthermore, a two-way ANOVA analysis indicated that both the wetland type and LULC had a significant (P<0.001) impact on TBC dynamics, individually and in combination.

3.3 Intra- and inter-wetland variations of LULC- specific TBC components

At the very outset, it is noteworthy to mention that, nine LULCs were identified within the restored wetland, whereas, only four LULCs were found in the degraded wetland. However, there were some common LULC categories within both of the wetlands. Hence, the inter-wetland variations of TBC components were analyzed based on these four common LULC categories only.
In the restored wetland, dense mangrove (µ: 164.36 Mg ha-1, σ: ±151.36 Mg ha-1) recorded highest AGB density, followed by the areas covered by mixed vegetation (µ: 102.66 Mg ha-1, σ: ±114.59 Mg ha-1), and agriculture (µ: 74.57 Mg ha-1, σ: ±79.01 Mg ha-1). Although, the lowermost AGB density was obtained in aquaculture (µ: 5.02 Mg ha-1, σ: ±0.27 Mg ha-1) category. In degraded wetland, a high concentration of AGB density was found under open mangrove (µ: 95.73 Mg ha-1, σ: ±60.18 Mg ha-1), while the lowermost AGB density was found in waterbody (µ: 6.01 Mg ha-1, σ: ±1.60 Mg ha-1). Notably, the average AGB density of open mangrove category was considerably higher in degraded wetland than that of restored wetland (Fig. 4). The amount of AGB also differed considerably under bare earth and sand category, where the amount of AGB was much higher in the restored wetland (µ: 36.71 Mg ha-1, σ: ± 47.95 Mg ha-1) than in the degraded wetland (µ: 11.80 Mg ha-1, σ: ±8.20 Mg ha-1). Although the mean AGB densities of herbaceous vegetation and water bodies were comparatively greater at the restored site than that at the degraded site, those differences were negligible. The mean AGB densities under different LULC categories significantly differed (P< 0.05). The only exceptions were with respect to aquaculture and waterbody in the restored wetland as well as between herbaceous vegetation and bare earth and sand in the de- graded wetland.
Fig. 4 Variability of TBC components among different LULC classes of restored and degraded wetlands

Note: Dense mangrove (µ: 306.88 Mg ha-1, σ: ±106.52 Mg ha-1) recorded maximum SOC density, followed by open mangrove (µ: 152.56 Mg ha-1, σ: ±95.10 Mg ha-1) and waterbody (µ: 107.44 Mg ha-1, σ: ±113.43 Mg ha-1) categories of the restored wetland (Table 2). In contrast, moderately low SOC densities were observed for the remaining LULCs. In the degraded wetland, the maximum SOC density was detected in bare earth and sand (µ: 102.02 Mg ha-1, σ: ±35.15 Mg ha-1), whereas the rest of the LULCs exhibited moderate-low SOC density. Furthermore, it's worth noting that in the restored wetland, open mangrove, herbaceous vegetation, and water bodies exhibited noticeably higher mean SOC densities compared to their counterparts. Conversely, the SOC density of bare earth and sand was found to be relatively higher in the degraded wetland compared to the restored wetland. The average SOC densities exhibited significant variations (P<0.05) across LULC categories of the restored wetland, except between bare earth and sand and Casuarina plantation. In contrast, mean SOC density displayed significant variations among all LULC categories in the degraded wetland. Additionally, the results of the two-way ANOVA indicated that wetland type and LULC have a significant (P<0.001) impact on AGB, BGB, SOC, and TBC dynamics, both independently and in combination.

Table 2 Estimated mean (μ), standard deviation (σ), and modal class values of AGB, BGB, and SOC densities under different LULC categories
Wetland LULC category AGB (Mg ha-1) BGB (Mg ha-1) SOC (Mg ha-1)
μ±σ Modal class μ±σ Modal class μ±σ Modal class
Restored Dense mangrove 164.36±151.36A >150 72.32±80.06A <30 306.88±106.52A >150
Open mangrove 51.20±50.52B <30 15.63±19.95B <30 152.56±95.10B >150
Mixed vegetation 102.66±114.59C <30 23.25±24.76C <30 41.17±43.31C <30
Casuarina plantation 21.20±26.81D <30 5.22±6.10D <30 54.90±57.99D <30
Herbaceous vegetation 16.88±25.48E <30 4.04±8.47D <30 89.52±77.14E <30
Agriculture 74.57±79.01F <30 17.20±17.38B <30 42.60±42.08F <30
Aquaculture 5.02±0.27G <30 1.36±0.07E <30 72.15±73.54G 30-60
Bare earth and sand 36.71±47.95H <30 8.74±10.72F <30 56.93±61.86D <30
Waterbody 8.92±17.27G <30 2.27±4.00E <30 107.44±113.43H <30
Degraded Open mangrove 95.73±60.18A >150 32.61±25.69A <30 44.47±20.37A 30-60
Herbaceous vegetation 11.80±8.96B <30 2.06±2.52B <30 77.24±41.16B 60-90
Bare earth and sand 11.23±8.20B <30 2.90±1.96B <30 102.02±35.15C 90-120
Waterbody 6.01±1.60C <30 1.61±0.40B <30 84.16±39.32D 90-120

Note: Diverse superscript letters show significant deviations for a specific parameter across LULC categories as per one-way ANOVA (F) first and thereafter through Tukey’s HSD Post Hoc measure (P<0.05). Degrees of freedom (df) are (8, 3376282) for restored and (3, 211381) for degraded wetlands, respectively.

The intra- and inter-wetland distribution of LULC-specific BGB density follows an identical pattern similar to AGB distribution, as it was derived from AGB itself. However, in the restored wetland, no statistically significant variation (P<0.05) of average BGB densities among open mangrove and agriculture, Casuarina plantation and herbaceous vegetation, as well as aquaculture and waterbody was found. Similarly, in the degraded wetland, no significant deviation (P<0.05) were recorded in average BGB densities among herbaceous vegetation, bare earth and sand, and waterbody.

4 Discussion

LULC is a major indicator of ecosystem health, biodiversity, and C sequestration potential (Hoque et. al., 2021). Satellite data-based LULC pattern identification, in this regard, helps towards conservation policy framework to enhance the C sequestration potential of various LULCs and ecosystems in the present global climate scenario (Ma et al., 2019; Mengist et al., 2023). Generally, wetlands, particularly those adorned with vegetation covers like mangroves, exhibit notably greater capacity for C sequestration with respect to other LULC classes. Therefore, these areas should be accorded the utmost priority in terms of regular monitoring, conservation efforts, and restoration initiatives (Sahu et al., 2016; Sharma et al., 2020; Hoque et al., 2021).
Although AGB constitutes a vital component of TBC, it performs a crucial role in driving the dynamics of both BGB and SOC of an ecosystem. The distributional pattern of AGB varies with the LULC pattern, depending upon the existence of vegetal cover and biophysical attributes of the prevailing flora. However, the variability of these biophysical attributes chiefly hinges on the type and health of the specific ecosystem. Generally, areas endowed with dense vegetation cover tend to exhibit higher AGB density (Zheng et al., 2004). This observation is in conformity with the findings of our current investigation, where it was found that AGB was highest in dense mangrove (µ: 164.36 Mg ha-1, σ: ±151.36 Mg ha-1) and mixed vegetation (µ: 102.66 Mg ha-1, σ: ±114.59 Mg ha-1) in the restored wetland and under the open mangrove in the degraded wetland (µ: 95.73 Mg ha-1, σ: ±60.18 Mg ha-1). Mean AGB density (µ: 51.20 Mg ha-1, σ: ±50.52 Mg ha-1) of open mangrove in the restored wetland was comparatively lower because this open mangrove category mostly consists of plantations and young stands mangrove vegetation. Consequently, the remaining LULCs in both wetlands exhibited comparatively lower AGB densities, primarily due to the absence or limited extent of vegetative cover.
SOC stocks varied over various LULC categories, as the sequestration quantity, nature of decomposition, and fixation vary over various LULC categories (Dorji et al., 2014). In the restored wetland, dense mangrove (µ: 306.88 Mg ha-1, σ: ±106.52 Mg ha-1) and open mangrove (µ: 152.56 Mg ha-1, σ: ±95.10 Mg ha-1) classes recorded high SOC density, owing to the accumulation of substantial amount of AGB as well as BGB under anaerobic conditions over an extended period (Datta et al., 2022). Conversely, even though the mixed vegetation and Casuarina plantation classes had substantial inputs of aboveground and belowground biomass, their SOC densities were comparatively lower. This discrepancy can be explained by the fact that these classes are situated on dunes along the shoreline and inland areas. In the degraded wetland, open mangrove (µ: 95.73 Mg ha-1, σ: ±60.18 Mg ha-1) exhibited low SOC density, indicating degradation in the open mangrove areas of this particular wetland ecosystem. This degradation may lead to the atmospheric release of previously sequestered SOC (Macreadie et al., 2013; Roy et al., 2021). Conversely, high SOC densities in bare earth and sand (µ: 102.02 Mg ha-1, σ: ±35.15 Mg ha-1) categories was observed in the degraded wetland, which probably indicates recent deforestation in areas where vegetation cover existed in the past (Datta et al., 2022). A specific inference could be made from this research that the TBC density of the restored wetland (µ: 182.68 Mg ha-1) is substantially greater than that of the degraded wetland (µ: 93.03 Mg ha-1). This disparity is a result of the combined effects of AGB, BGB, and SOC dynamics across various LULCs and wetlands. The variabilities of TBC and its components (AGB, BGB, and SOC) of a particular LULC category across wetlands highlight the impact of wetland health on C dynamics. Additionally, this variation manifests the singular and joint impacts of wetland and LULC categories on blue C dynamics, which was also evident from the two-way ANOVA.

5 Conclusions

The current investigation furnished a comprehensive analysis of TBC dynamics, including its constituent components viz., AGB, BGB, and SOC. It assessed the influence of LULC categories and wetland type on the heterogeneity of these C stocks in restored and degraded wetlands of MCP, India. Firstly, the study unveiled a consistent pattern of higher densities of blue C components in the dense mangrove and open mangrove categories, whereas the remaining LULCs displayed moderate to low density in the restored wetland. Conversely, in the degraded wetland, high densities of carbon pools were observed in the open mangrove category, while herbaceous vegetation, bare earth and sand, and water bodies exhibited lower densities. However, there were some noteworthy exceptions. In the restored wetland, open mangrove displayed low biomass densities, primarily stemming from the prevalence of plantations and young mangrove stands in this particular LULC category. Subsequently, in the degraded wetland, bare earth and sand exhibited high SOC density, suggesting a recent deforestation event within this wetland area. Secondly, the estimated TBC stock of the restored wetland (246710.91 Mg, µ: 182.68 Mg ha-1) was substantially greater with respect to the degraded wetland (7865.49 Mg, µ: 93.03 Mg ha-1). Thirdly, the mean AGB, BGB, SOC, and TBC densities of most of the LULC categories significantly (P<0.05) differed in both wetlands. It was also evident from this study that LULC and wetland type, individually and in combination, have a noteworthy (P<0.001) influence on blue C pools. Cumulatively, findings of this investigation can be employed as initial reference points which will aid in policy making decisions regarding restoration, conservation, and management of these wetlands as efficient blue C pools, highlighting their importance as a key element of mitigating climate change.

Acknowledgments

The authors recognize the enormous help rendered by the local populace of Medinipur Coastal Plain during field investigations.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Ali A, Xu M S, Zhao Y T, et al. 2015. Allometric biomass equations for shrub and small tree species in subtropical China. Silva Fennica, 49(4): 1-10.

[2]
Babu K N, Mandyam S, Jetty S, et al. 2023. Carbon stocks of tree plantations in a Western Ghats landscape, India: Influencing factors and management implications. Environmental Monitoring and Assessment, 195(3): 404. DOI: 10.1007/s10661-023-10964-w.

PMID

[3]
Bassi N, Kumar M D, Sharma A, et al. 2014. Status of wetlands in India: A review of extent, ecosystem benefits, threats and management strategies. Journal of Hydrology: Regional Studies, 2: 1-19.

[4]
Bazzo C O G, Kamali B, Hütt C, et al. 2023. A review of estimation methods for aboveground biomass in grasslands using UAV. Remote Sensing, 15(3): 639. DOI: 10.3390/rs15030639.

[5]
Bhomia R K, MacKenzie R A, Murdiyarso D, et al. 2016. Impacts of land use on Indian mangrove forest carbon stocks: Implications for conservation and management. Ecological Applications, 26(5): 1396-1408.

DOI PMID

[6]
Chave J, Andalo C, Brown S, et al. 2005. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia, 145: 87-99.

DOI PMID

[7]
Cummings D L, Kauffman J B, Perry D A, et al. 2002. Aboveground biomass and structure of rainforests in the southwestern Brazilian Amazon. Forest Ecology and Management, 163(1-3): 293-307.

[8]
Das S, Deb S, Banik B, et al. 2021. Estimation of soil organic carbon pools and biomass carbon stocks in different aged rubber (Hevea brasiliensis muell. arg.) plantations of Tripura. Journal of the Indian Society of Soil Science, 69(3): 248-260.

[9]
Dash P, Sanders S L, Parajuli P, et al. 2023. Improving the accuracy of land use and land cover classification of Landsat data in an agricultural watershed. Remote Sensing, 15(16): 4020. DOI: 10.3390/rs15164020.

[10]
Datta D, Bairagi M, Dey M, et al. 2022. Spatially explicit estimation of soil organic carbon stock of an estuarine mangrove wetland of eastern India using elemental analysis and very-fine resolution satellite data. Ecological Processes, 11: 1-16.

[11]
Datta D, Deb S. 2017. Forest structure and soil properties of mangrove ecosystems under different management scenarios: Experiences from the intensely humanized landscape of Indian Sunderbans. Ocean and Coastal Management, 140: 22-33.

[12]
Datta D, Dey M, Ghosh P, et al. 2023. Coupling multi-sensory earth observation datasets, in-situ measurements, and machine learning algorithms for total blue C stock estimation of an estuarine mangrove forest. Forest Ecology and Management, 546: 1-10.

[13]
Datta D, Roy A K, Kundu A, et al. 2021. An alternative approach to delineate wetland influence zone of a tropical intertidal mudflat using geoinformation technology. Estuarine, Coastal and Shelf Science, 253: 1-14.

[14]
Deb D, Singh J P, Deb S, et al. 2017. An alternative approach for estimating above ground biomass using Resourcesat-2 satellite data and artificial neural network in Bundelkhand region of India. Environmental Monitoring and Assessment, 189: 1-12.

[15]
Dorji T, Odeh I O A, Field D J, et al. 2014. Digital soil mapping of soil organic carbon stocks under different land use and land cover types in montane ecosystems, Eastern Himalayas. Forest Ecology and Management, 318: 91-102.

[16]
Dung L V, Tue N T, Nhuan M T, et al. 2016. Carbon storage in a restored mangrove forest in Can Gio Mangrove Forest Park, Mekong Delta, Vietnam. Forest Ecology and Management, 380: 31-40.

[17]
Fourqurean J, Johnson B, Kauffman J B, et al. 2014. Field sampling of soil carbon pools in coastal ecosystems. In: Howard J, Hoyt S, Isensee K, et al (eds.). Coastal blue carbon: Methods for assessing carbon stock and emissions factors in mangroves, tidal salt marshes, and seagrasses. Arlington, USA: Conservation International, Intergovernmental Oceanographic Commission of UNESCO, International Union for Conservation of Nature.

[18]
Hamzeh M A, Lahijani H A K. 2022. Soil and vegetative carbon sequestration in Khuran Estuary Mangroves, Strait of Hormoz, during the last 18 centuries. Estuaries and Coasts, 45: 1583-1595.

[19]
Haywood B J, Hayes M P, White J R, et al. 2020. Potential fate of wetland soil carbon in a deltaic coastal wetland subjected to high relative sea level rise. Science of the Total Environment, 711: 135185. DOI: 10.1016/j.scitotenv.2019.135185.

[20]
Hoque M Z, Cui S, Islam I, et al. 2020. Future impact of land use/land cover changes on ecosystem services in the lower meghna river estuary, Bangladesh. Sustainability, 12(5): 2112. DOI: 10.3390/su12052112.

[21]
Hoque M Z, Cui S, Islam I, et al. 2021. Dynamics of plantation forest development and ecosystem carbon storage change in coastal Bangladesh. Ecological Indicators, 130: 107954. DOI: 10.1016/j.ecolind.2021.107954.

[22]
Hutchinson J, Manica A, Swetnam R, et al. 2014. Predicting global patterns in mangrove forest biomass. Conservation Letters, 7(3): 233- 240.

[23]
Jia K, Wei X, Gu X, et al. 2014. Land cover classification using Landsat 8 operational land imager data in Beijing, China. Geocarto International, 29(8): 941-951.

[24]
Kauffman J B, Cole T G. 2010. Micronesian mangrove forest structure and tree responses to a severe typhoon. Wetlands, 30: 1077-1084.

[25]
Kauffman J B, Donato D C. 2012. Protocols for the measurement, monitoring and reporting of structure, biomass and carbon stocks in mangrove forests. Bogor, Indonesia: Center for International Forestry Research.

[26]
Kenzo T, Yoneda R, Alias M A. 2023. Estimation of above and belowground biomass for grass, herb, and fern species in Peninsula Malaysia. Tropical Ecology, 64(3): 500-510.

[27]
Lu D. 2005. Aboveground biomass estimation using Landsat TM data in the Brazilian Amazon. International Journal of Remote Sensing, 26(12): 2509-2525.

[28]
Luo S, Wang C, Xi X, et al. 2017. Fusion of airborne LiDAR data and hyperspectral imagery for aboveground and belowground forest biomass estimation. Ecological Indicators, 73: 378-387.

[29]
Luo Y, Wang X, Zhang X, et al. 2012. Root: shoot ratios across China’s forests: Forest type and climatic effects. Forest Ecology and Management, 269: 19-25.

[30]
Ma T, Li X, Bai J, et al. 2019. Four decades’ dynamics of coastal blue carbon storage driven by land use/land cover transformation under natural and anthropogenic processes in the Yellow River Delta, China. Science of the Total Environment, 655: 741-750.

[31]
Macreadie P I, Hughes A R, Kimbro D L. 2013. Loss of ‘Blue Carbon’ from coastal salt marshes following habitat disturbance. Plos One, 8(7): e69244. DOI: 10.1371/journal.pone.0069244.

[32]
Madsen J D. 1993. Biomass techniques for monitoring and assessing control of aquatic vegetation. Lake and Reservoir Management, 7(2): 141- 154.

[33]
Manuri S, Brack C, Nugroho N P, et al. 2014. Tree biomass equations for tropical peat swamp forest ecosystems in Indonesia. Forest Ecology and Management, 334: 241-253.

[34]
McLeod E, Chmura G L, Bouillon S, et al. 2011. A blueprint for blue carbon: Toward an improved understanding of the role of vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment, 9(10): 552-560.

[35]
Mengist W, Soromessa T, Feyisa G L. 2023. Responses of carbon sequestration service for landscape dynamics in the Kaffa biosphere reserve, southwest Ethiopia. Environmental Impact Assessment Review, 98: 106960. DOI: 10.1016/j.eiar.2022.106960.

[36]
Mohanta M R, Mohanta A, Mohapatra U, et al. 2020. Carbon stock assessment and its relation with tree biodiversity in tropical moist deciduous forest of similipal biosphere reserve, Odisha, India. Tropical Ecology, 61: 497-508.

[37]
Nandi G, Neogy S, Roy A K, et al. 2020. Immediate disturbances induced by tropical cyclone Fani on the coastal forest landscape of eastern India: a geospatial analysis. Remote Sensing Applications: Society and Environment, 20: 100407. DOI: 10.1016/j.rsase.2020.100407.

[38]
OFSDP Odisha Forestry Sector Development Project. 2010. Orissa Forestry Sector Development Project Newsletter. http://www.ofsds.in/Publication/Dec2010E.pdf. Viewed on 2023-06-28.

[39]
Peichl M, Arain M A. 2007. Allometry and partitioning of above and belowground tree biomass in an age-sequence of white pine forests. Forest Ecology and Management, 253(1): 68-80.

[40]
Pena M A, Brenning A. 2015. Assessing fruit-tree crop classification from Landsat-8 time series for the Maipo Valley, Chile. Remote Sensing of Environment, 171: 234-244.

[41]
Pendleton L, Donato D C, Murray B C, et al. 2012. Estimating global “Blue Carbon” emissions from conversion and degradation of vegetated coastal ecosystems. Plos One, 7(9): e43542. DOI: 10.1371/journal.pone.0043542.

[42]
Perera N, Lokupitiya E, Halwatura D, et al. 2022. Quantification of blue carbon in tropical salt marshes and their role in climate change mitigation. Science of the Total Environment, 820: 153313. DOI: 10.1016/j.scitotenv.2022.153313.

[43]
Rani V, Nandan S B, Schwing P T. 2021. Carbon source characterisation and historical carbon burial in three mangrove ecosystems on the south west coast of India. Catena, 197: 104980. DOI: 10.1016/j.catena.2020.104980.

[44]
Roy A K, Datta D. 2018. Analyzing the effects of afforestation on estuarine environment of river Subarnarekha, India using geospatial technologies and participatory appraisals. Environmental Monitoring and Assessment, 190: 645. DOI: 10.1007/s10661-018-7030-7.

PMID

[45]
Roy A K, Ghosh P K, Datta D. 2020. Typological inventorization and rapid ecological health assessment of the wetlands of Medinipur coastal plain, India. Folia Geographica, 62(2): 22-47.

[46]
Roy A K, Naskar M, Chakraborty T, et al. 2021. Assessing the impact of aquafarming on landscape dynamics of coastal West Bengal, India using remotely sensed data and spatial metrics. In: Rani M, Seenipandi K, Rehman S, et al (eds). Remote Sensing of Ocean and Coastal Environments. Amsterdam, The Netherlands: Elsevier: 117-38.

[47]
Sahu S C, Kumar M, Ravindranath N H. 2016. Carbon stocks in natural and planted mangrove forests of Mahanadi Mangrove Wetland, East Coast of India. Current Science, 110(12): 2253-2260.

[48]
Senger D F, Hortua D S, Engel S, et al. 2021. Impacts of wetland dieback on carbon dynamics: A comparison between intact and degraded mangroves. Science of the Total Environment, 753: 141817. DOI: 10.1016/j.scitotenv.2020.141817.

[49]
Sharma S, MacKenzie R A, Tieng T, et al. 2020. The impacts of degradation, deforestation and restoration on mangrove ecosystem carbon stocks across Cambodia. Science of the Total Environment, 706: 135416. DOI: 10.1016/j.scitotenv.2019.135416.

[50]
Sutton-Grier A E, Sandifer P A, 2019. Conservation of wetlands and other coastal ecosystems: A commentary on their value to protect biodiversity, reduce disaster impacts, and promote human health and well-being. Wetlands, 39(6): 1295-1302.

DOI

[51]
UNEP (United Nations Environment Programme) and CIFOR (Center for International Forestry Research). 2014. Guiding principles for delivering coastal wetland carbon projects. Nairobi, Kenya: UNEP and Bogor, Indonesia: CIFOR.

[52]
Vågen T G, Winowiecki L A. 2013. Mapping of soil organic carbon stocks for spatially explicit assessments of climate change mitigation potential. Environmental Research Letters, 8(1): 015011. DOI: 10.1088/1748-9326/8/1/015011.

[53]
Virgulino-Júnior P C C, Carneiro D N, Nascimento Jr W R, et al. 2020. Biomass and carbon estimation for scrub mangrove forests and examination of their allometric associated uncertainties. Plos One, 15(3): 1-18.

[54]
WAC World Agroforestry Centre. 2011. World Agroforestry Centre databases. http://www.worldagroforestrycentre.org/our_products/databases/. Viewed on 2023-06-30.

[55]
Wu W, Zhang J, Bai Y, et al. 2023. Aboveground biomass dynamics of a coastal wetland ecosystem driven by land use/land cover transformation. Remote Sensing, 15(16): 3966. DOI: 10.3390/rs15163966.

[56]
Xi Y, Peng S, Ciais P, et al. 2021. Future impacts of climate change on inland Ramsar wetlands. Nature Climate Change, 11: 45-51.

[57]
Yang R M, Huang L M, Zhang X, et al. 2023. Mapping the distribution, trends, and drivers of soil organic carbon in China from 1982 to 2019. Geoderma, 429: 116232. DOI: 10.1016/j.geoderma.2022.116232.

[58]
Zheng D, Rademacher J, Chen J, et al. 2004. Estimating aboveground biomass using Landsat 7 ETM+ data across a managed landscape in northern Wisconsin, USA. Remote Sensing of Environment, 93(3): 402- 411.

[59]
Zhou X, Yin Z, Zhou Y, et al. 2023. Modeling stand biomass for Moso bamboo forests in Eastern China. Frontiers in Plant Science, 14: 1186250. DOI: 10.3389/fpls.2023.1186250.

Options
Outlines

/

〈 〉